1. Field of the Invention
The present invention relates to an iris recognition method and system using the same, and more particularly, to an iris recognition method and system using the same to improve a success ratio and a speed of iris recognition by registering iris code for each of illumination locations.
2. Description of the Related Art
As known, further to the conventional contact type or non-contact type card systems for security, crime prevention or authentication of identification, systems which allow or deny the entrance to a specific area or the access to specific information according to authentication performed by recognizing finger print or iris, have come to wide use.
The iris recognition system has advantages, such as higher and more precise recognition rate compared with the finger print recognition system and impossibility in forgery and stealing since the iris pattern is different every human being and there are amazingly a lot of kinds of the iris patterns.
In general, the iris has a characteristic in that the pattern thereof does not vary for a long time. In 1980s, American Ophthalmologists, e.g., Leonard Flom and Alan Safir found that every human being has his (her) inherent iris patterns. A basic patent of the technology related to iris recognition was registered in 1987. A professor, John G. Daugman of University of Cambridge of the United Kingdom proposed an image processing algorithm based on Gabor wavelet transform in which the iris pattern can be coded in 512 bytes. The currently commercialized products are based on this algorithm.
An initial available iris recognition system was developed in 1995 by IriScan, Inc. founded around core members of the aforementioned three peoples in the United States. The iris recognition system includes an image capture unit for capturing an eye image and a recognition unit for recognizing a person using the captured eye image. The image capture unit functions to capture an eye image suitable for iris recognition and the recognition unit performs functions, such as iris region extraction, iris feature extraction, registration and recognition from the captured eye image. Accordingly, the iris recognition system is employed by banks that require a high security. Further, it is applied to entrance doors of buildings and personal computers to prevent personal information from leaking.
FIG. 1 illustrates a general iris recognition system. Referring to FIG. 1, when a user accesses to an iris recognition system, a distance measure sensor 109 measures a distance from the user and sends the measured distance to the a control device 105 through a driver 107. The control device 105 determines whether the inputted distance measurement value is within a predetermined operation range. The control device 105 sends a control signal to the driver 107 if the inputted distance measurement value is within a predetermined operation range. The driver 107 sends an active signal to an external indicator 108 to inform a user of the fact that system is operating by generating an indication. Here, the external indicator 108 can be a light emission device such as an LED. When a user positions his or her eye on light axis of a camera 103 through an optical window 101, a cold mirror 102 screens a visible ray and transmits an infrared ray. The location at which an iris should be positioned is indicated for the user to ascertain whether his or her eye is located on the light axis of the camera 103. An illuminating device controls illumination intensity. Here, the control device 105 controls the camera 103 to zoom-in, zoom-out, and focus using the distance measurement value measured deliberately. The control device 105 sends a camera control signal to the camera 103 to capture an iris image from the user's eye. The iris image captured by the camera 103 is sent to a frame grabber 104 and the frame grabber 104 generates an iris code from the iris image. Here, the iris code implies IrisCode and an iris recognition algorithm separates its brightness pattern into eight circles. The eight circles are obtained by analysis on each area. The control device 105 registers this iris code on the iris code database 110 and determines authentication of the user using the authentication code. The performance of such an iris recognition system depends on how fast and how precisely to recognize the iris. For this, images are captured changing the illumination location to avoid second reflection caused by glasses using three illuminators (for example, LEDs).
FIG. 2 illustrates configuration of an illuminating device that has a plurality of illuminators in the iris recognition system in the prior art. Referring to FIG. 2, when an illuminating device has a plurality of illuminators and a user wears a pair of glasses, reflection can occur from iris image according to the angles between a pair of glasses of the user and three illuminators 204, 205 and 206. Accordingly, an image processor 202 detects reflection degree and sends glasses reflection information to an illumination controller 203 prior to calculating focus degree.
The illumination controller 203 controls switching (On/Off) of three illuminators 204, 205 and 206 based on the glasses reflection information to avoid glasses reflection. The reflection locations of the glasses are changed according to selective On/Off of the three illuminators 204, 205 and 206. In other words, after only the illuminator 204 is turned on, an iris image is captured from a user's iris. After that, an iris code is generated from the iris image and then recognition is performed. In the above case, when recognition is failed due to the reflection of glasses, the illuminator 204 is turned off and another illuminator 205 at different location is turned on. An iris image is again captured from a user's iris. An iris code is generated from the iris image and then recognition is tried. This procedure is repeatedly performed with changing the illuminators until the recognition succeed in.
However, when the recognition is performed repeatedly using the illuminating device described above, recognition speed gets slow down and recognition success ratio is lowered.